Run 4. Outer Step 52. Inner Step 21.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5575626,
-    5575628,
-    5575631,
-    5575633,
-    5575634,
-    5575638,
-    5575640
+    5575644,
+    5575648,
+    5575651,
+    5575653,
+    5575655
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5574901,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0928d162b958916966271078800654c27efabbc6f6cbbb0953b3b76a804d01fd
+oid sha256:b7511f85b3e5dad4e306efaef8f147ec2a92bc8b2944e607ebd82bd3fb94757a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3481715f22f38b6ad1f8ba446830547821fd7c1f504f60fd458fd390b45cb53
+oid sha256:566294bb2a77bd7c3c7efb14351e96202ea42cadef3539d1cc2e004973d8a370
 size 4040701744